Accelerating Block Coordinate Descent for Nonnegative Tensor Factorization
Numerical Analysis
2021-05-12 v2 Machine Learning
Numerical Analysis
Optimization and Control
Machine Learning
Abstract
This paper is concerned with improving the empirical convergence speed of block-coordinate descent algorithms for approximate nonnegative tensor factorization (NTF). We propose an extrapolation strategy in-between block updates, referred to as heuristic extrapolation with restarts (HER). HER significantly accelerates the empirical convergence speed of most existing block-coordinate algorithms for dense NTF, in particular for challenging computational scenarios, while requiring a negligible additional computational budget.
Cite
@article{arxiv.2001.04321,
title = {Accelerating Block Coordinate Descent for Nonnegative Tensor Factorization},
author = {Andersen Man Shun Ang and Jeremy E. Cohen and Nicolas Gillis and Le Thi Khanh Hien},
journal= {arXiv preprint arXiv:2001.04321},
year = {2021}
}
Comments
32 pages, 24 figures